creating fields with awk

[rog64]

Hi, i am writing a bash script and i am trying to redirect output to specific fields in a file. I cant seem to find any way of doing this. I have tried to create fields and add to a specific field using the following line
printf "%s\t%s\t%s\t%s\t%s\n" "" "" "$total" "" "" >> study.txt
which works, but when i try to retrieve info from a specific field using the following line:
cat study.txt | awk '{ sum += $1} END {print sum}'
it doesnt give me the sum of field 1, it gives the sum of all fields combined. Can you add to a specific field using awk?

 

[feherke]

Hi

[pre] In the special
case that FS is a single space, fields are separated by runs of spaces
and/or tabs and/or newlines.
[/pre](...)[pre]

FS The input field separator, a space by default.[/pre]

So your multiple adjacent [tt]\t[/tt] characters are counted as a single one, making the $total you output to be the 1^st field, not the 3^rd.

Not sure what is your ultimate goal, but looks like you need to explicitly specify the field separator :

awk [highlight]-F '\t'[/highlight] '[teal]{[/teal] sum [teal]+=[/teal] [navy]$1[/navy][teal]}[/teal] [b]END[/b] [teal]{[/teal][b]print[/b] sum[teal]}[/teal]'

Feherke.
feherke.ga

 

[rog64]

Thanks for the reply. I am not using the printf line i entered above. it does not work. I am trying to create five fields seperated by tabs, once this is done i need to enter info into each field seperately. Can this be done using awk?

 

[feherke]

Hi

AWK processes data as a stream. Although that stream can come from a file, AWK itself not handles it as file. From your last post I understand you want something like in-place editing of a file. AWK is not able to do that. The closest alternative would be Perl.

For example, having a file with 3 rows * 5 fields, this is how you place a value in 2^nd row 3^rd field :

[blue]master #[/blue] echo $'\t\t\t\t\n\t\t\t\t\n\t\t\t\t' > rog64

[blue]master #[/blue] tr '\t' '|' < rog64 
||||
||||
||||

[gray]#                      ,---------------------------------- input field separator[/gray]
[gray]#                      |         ,------------------------ field number, counted from 0[/gray]
[gray]#                      |         |  ,--------------------- new value[/gray]
[gray]#                      |         |  |       ,------------- row number, counted from 1[/gray]
[gray]#                      |         |  |       |          ,-- output field separator[/gray]
[blue]master #[/blue] perl -pi -aF'\t' -e '[navy]$F[/navy][teal][[/teal][purple]2[/purple][teal]]=[/teal][purple]42[/purple][b]if[/b][navy]$.[/navy][teal]==[/teal][purple]2[/purple][teal];[/teal][navy]$_[/navy][teal]=[/teal][b]join[/b][i][green]"\t"[/green][/i][teal],[/teal][navy]@F[/navy]' rog64

[blue]master #[/blue] tr '\t' '|' < rog64 
||||
||42||
||||

Feherke.
feherke.ga

 

[feherke]

Hi

By the way, for so extremely simple thing like replacing one field, Sed could be also enough :

[blue]master #[/blue] echo $'\t\t\t\t\n\t\t\t\t\n\t\t\t\t' > rog64

[gray]#                ,--------------- row number, counted from 1[/gray]
[gray]#                |     ,--------- field separator[/gray]
[gray]#                |     |   ,----- new value[/gray]
[gray]#                |     |   |  ,-- field number, counted from 1[/gray]
[blue]master #[/blue] sed -i '2s/[^\t]*/42/3' rog64 

[blue]master #[/blue] tr '\t' '|' < rog64 
||||
||42||
||||

Feherke.
feherke.ga

 

[rog64]

Wow, great response and well explained, thanks alot, it really helped.